Project 2, "Risk Factors for Fracture Among the Elderly," complements the pathophysiology studies in this Program Project Grant by providing a population perspective on risk factors for, and consequences of, osteoporotic fractures. We propose to continue our focus on better identifying risk factors for long-term fracture outcomes by conducting large historical cohort studies of Olmsted County, MN residents with specific conditions that exacerbate bone loss from estrogen deficiency and aging ("secondary" osteoporosis). We also will provide important updated information on fracture incidence trends and cost, data which have been essential for quantifying the public health burden of osteoporosis nationally. These studies are possible due to the unique resources of the Rochester Epidemiology Project, a medical records linkage system which provides access to detailed medical data for an entire geographically-defined population, and the Olmsted County Healthcare Expenditure and Utilization Database, which estimates the cost of all outpatient and inpatient medical care delivered to local residents. Our choice of aims for this funding cycle is especially relevant in light of the new World Health Organization approach to fracture risk assessment, which identifies rheumatoid arthritis (RA) and personal fracture history as key risk factors. Since our preliminary data suggest that most RA patients will not fracture over their remaining lifetimes, we not only will characterize high risk individuals, but also those at low risk for fracture who might be spared costly osteoporosis treatment. Furthermore, fractures occurring before age 40 years are presently ignored in osteoporosis risk assessment so we will use novel technologies to determine if those who suffered a forearm fracture in childhood (the most common fracture in youth) have deficits in bone metabolism, structure, or strength that persist into adulthood, as some have suggested, and whether they are at increased risk for fractures in later life. The descriptive incidence and cost data for fractures at all skeletal sites are needed for related policy determinations (e.g., cost-effectiveness of fracture prophylaxis). The population-based studies proposed will therefore fill important gaps in our knowledge of age-related fracture risk. RELEVANCE (See instructions): Because the aging population is increasingly at risk of devastating osteoporotic fractures, it is important to identify high-risk individuals for more aggressive osteoporosis management, while also characterizing lowrisk individuals who might appropriately be spared treatment expense and side effects. Updated information on fracture frequency and cost is needed to help define the most efficient approaches to fracture prevention.